Most research work on news mining nowadays covers phrase and topic level. A few works conducted on logical level mainly focus on personalized news service and no special efforts are put on the applications of ontology techniques on deep news mining. In this paper, we demonstrate a whole strategy for deeply understanding event-focused news taking the advantage of ontology representation and ontology reasoning. We propose an ontology-enriched news deep understanding framework ONDU which addresses the following problems: (1) how to transfer parsed news content into logical triples by using domain ontology. (2) The application of ONDU based on the reasoning results from the ontology reasoner TrOWL over the RDF data expressing the news. Through this whole strategy we can detect the inconsistence among multiple news articles and compare the different information implied in different news. We can even integrate a set of news content through merging the RDF data. The empirical experiment conducted on news from several portals shows the effectiveness and usefulness of our method.
|Name||Communications in Computer and Information Science|
|Conference||7th Chinese Semantic Web Symposium and the 2nd Chinese Web ScienceConference, CSWS 2013|
|Period||12/08/13 → 16/08/13|
- News mining
- Text understanding